LED: Light Enhanced Depth Estimation at Night


Simon de Moreau (Mines Paris - PSL University), Yasser Almehio (Valeo AI), Andrei Bursuc (Valeo AI), Hafid EL IDRISSI (Valeo AI), Bogdan Stanciulescu (Mines Paris - PSL University), Fabien Moutarde (Mines Paris - PSL University)
The 35th British Machine Vision Conference
PDFPosterVideo (Right click to download)Supplementary

Abstract

Nighttime camera-based depth estimation is a highly challenging task, especially for autonomous driving applications, where accurate depth perception is essential for ensuring safe navigation. Models trained on daytime data often fail in the absence of precise but costly LiDAR. Even vision foundation models trained on large amounts of data are unreliable in low-light conditions. In this work, we aim to improve the reliability of perception systems at night time. To this end, we introduce Light Enhanced Depth (LED), a novel, cost-effective approach that significantly improves depth estimation in low-light environments by harnessing a pattern projected by high definition headlights available in modern vehicles. LED leads to significant performance boosts across multiple depth-estimation architectures (encoder-decoder, Adabins, DepthFormer, Depth Anything V2) both on synthetic and real datasets. Furthermore, increased performances beyond illuminated areas reveal a holistic enhancement in scene understanding. Finally, we release the Nighttime Synthetic Drive Dataset, a synthetic and photo-realistic nighttime dataset, which comprises 49,990 comprehensively annotated images. To facilitate further research, both synthetic dataset and code are publicly available at https://simondemoreau.github.io/LED/.

Citation

@inproceedings{Moreau_2025_BMVC,
author    = {Simon de Moreau and Yasser Almehio and Andrei Bursuc and Hafid EL IDRISSI and Bogdan Stanciulescu and Fabien Moutarde},
title     = {LED: Light Enhanced Depth Estimation at Night},
booktitle = {36th British Machine Vision Conference 2025, {BMVC} 2025, Sheffield, UK, November 24-27, 2025},
publisher = {BMVA},
year      = {2025},
url       = {https://bmva-archive.org.uk/bmvc/2025/assets/papers/Paper_1096/paper.pdf}
}


Copyright © 2025 The British Machine Vision Association and Society for Pattern Recognition
The British Machine Vision Conference is organised by The British Machine Vision Association and Society for Pattern Recognition. The Association is a Company limited by guarantee, No.2543446, and a non-profit-making body, registered in England and Wales as Charity No.1002307 (Registered Office: Dept. of Computer Science, Durham University, South Road, Durham, DH1 3LE, UK).

Imprint | Data Protection